1. Field of the Invention
The present application concerns method and apparatus for the generation of images. In particular the present application concerns method and apparatus for the generation of images of an object from a selected viewpoint from a plurality of images of the same object from other viewpoints.
2. Related Art
It is known to split two-dimensional image data into a number of frequency bands and process the different frequency band images separately. When generating images of three-dimensional objects from a number of different view of the same object, it would be desirable to be able to process image data for three dimensional objects in a similar way.
One way of extending the known processing of two-dimensional images would be to process images by:
projecting data from the available two-dimensional images on to the surface of an object;
blurring the generated textures across the surface of the three-dimensional object to obtain a low band textured surface;
subtracting the original textured surface to obtain the corresponding high band textured surface
and then blending textured surfaces using the different spacial frequency information for the different bands.
However, it is not known how to perform the blurring operation for textures on the surface of a model in an efficient and consistent way. An alternative method for generating spacial frequency image data for an image of a 3D object from a selected viewpoint is therefore required.